Structure and function of the nucleolus in the spotlight

The nucleolus is the most obvious and clearly differentiated nuclear sub-compartment. It is where ribosome biogenesis takes place, but it is becoming clear that the nucleolus also has non-ribosomal functions. In this review we discuss recent progress in our understanding of how both ribosome biosynthesis and some non-ribosomal functions relate to observable nucleolar structure. We still do not have detailed enough information about the in situ organization of the various processes taking place in the nucleolus. However, the present power of light and electron microscopy techniques means that a description of the organization of nucleolar processes at the molecular level is now achievable, and the time is ripe for such an effort.

Introduction

The nucleolus is the most distinctive nuclear sub-compartment (Figure 1, Figure 2) and is the site of the biogenesis of ribosomal RNA (rRNA) and 40S and 60S ribosomal subunits (r-subunits) [1, 2, 3, 4]. In metabolically active animal and plant somatic cells and in yeast, the nucleolus contains tens to hundreds of active r-genes, which account for about one-half of the total cellular RNA production. However, most of the evidence indicates that a significant fraction, often >50%, of r-genes remains transcriptionally silent [5].

Ribosomal genes comprise a transcribed sequence and an intergenic spacer, and are located on one or (usually) several chromosomes in arrays of head-to-tail tandem repeats called nucleolus organizer regions (NORs). For instance, human diploid cells contain ∼400 r-genes on five pairs of NOR-bearing chromosomes, each NOR containing several tens of repeats. During interphase, r-repeats from more than one NOR-bearing chromosome often cluster together and participate in the formation of a given nucleolus. In mitosis, rRNA synthesis ceases as a result of phosphorylation of the relevant nucleolar factors and the nucleoli disassemble; at the end of mitosis rRNA synthesis resumes and nucleoli reform [6•, 7, 8•, 9•, 10].

Ribosome biogenesis requires RNA polymerase I (pol I)-driven synthesis, cleavage and modification of precursor rRNAs (pre-rRNAs), assembly with r-proteins and transient interactions with numerous small nucleolar ribonucleoproteins (snoRNPs) and non-ribosomal nucleolar proteins, culminating in the export of almost mature r-subunits. Although this would appear to be a simple scenario, in fact the biogenesis of ribosomes is a very much more complex process involving many more steps and factors than was foreseen [2, 11, 12••].

In the past few years, our understanding of ribosome biosynthesis has been revolutionized by proteomic research backed by genetic and biochemical analyses [2], particularly in yeast. Many of the individual steps in ribosome biogenesis have now been described at the molecular level, and most of the macromolecular components involved have been identified. However, the synthesis of these data to understand the various interactions of the different nucleolar factors and r-proteins in vivo remains at an early stage, even in yeast [2, 11, 12••, 13, 14•, 15•, 16, 17•].

Proteomic research [12••, 18, 19, 20•, 21•] is also rapidly increasing our understanding of ribosome biosynthesis in higher eukaryotes, although at a slower pace than in yeast. It is important to emphasize that, although rRNA and r-proteins have been highly conserved among eukaryotes, it is not always possible to extrapolate yeast data to metazoa and other eukaryotes. At the molecular level, for example, the factors involved in the earliest steps of the pol I activation, the pre-initiation complex (PIC), differ between yeast and mammals [4, 5]. At the cellular level, the nucleolus in the yeast Saccharomyces cerevisiae does not disassemble at mitosis, in contrast to what occurs in mammals and plants. Furthermore, important findings with relevance to human medicine are very specific to mammalian or even to human cells [22•, 23•, 24].

It has also become clear that the nucleolar localization of many factors, even ‘typical’ nucleolar components, can be altered by cell cycle, modulation of cell growth or cell differentiation and, conversely, that many ‘non-nucleolar’ proteins or macromolecular complexes are often encountered in nucleoli [3, 25]. Different location of a macromolecule suggests a different function, and indeed there is much emerging evidence that nucleoli have a number of ‘non-canonical’ functions in addition to their roles as ribosome factories, such as virus infection control, maturation of non-nucleolar RNAs or RNPs, senescence and regulation of telomerase function, regulation of cell cycle, tumor suppressor and oncogene activities, cell stress sensing and signaling [3, 7, 25, 26, 27•, 28]. The best-characterised of these other activities is in the assembly of the signal recognition particle, a complex of an RNA with several proteins that targets translation of certain proteins to the endoplasmic reticulum [25, 29, 30, 31, 32]. This activity is relatively easily rationalized as being at least connected with ribosome biogenesis and activity.

In this review, we discuss recent progress in deciphering how the processes of the ribosome biosynthetic pathway are integrated into the nucleolar structure, with the emphasis on the important conceptual changes since the last structurally oriented nucleolar review published in this journal in 1999 [33]. It is impossible to survey such a vast literature adequately in a short review, and we have had to be very selective. For the most part we have concentrated where possible on human nucleoli, but have included data from other species, particularly yeast, where relevant, or where human data is still unavailable.